R workshop based on behavioral data

Program Overview (KST, GMT+9)

강사: 고언숙 (조선대학교), 마가렛 맥도날드 (University of Kansas), 채준호 (조선대학교)

Click here to join via Zoom

영한 번역 캡션이 자막으로 제공됩니다. Translation captions will be activated.

    Greetings and preparations

• 10:10 - 10:20

Welcoming remarks and announcements

• 10:20 - 10:30 강희숙 조선대 인문학연구원장; 고언숙 인문데이터과학연구소장

Morning session

• 10:30 - 11:00 | Research talk 1: 연속변수를 종속변수로 하는 연구 소개 (고언숙) 

Korean infants' perceptual responses to Korean and Western music based on musical experience

• 11:00 - 12:00 | Tutorial Session 1:선형 혼합 효과 회귀 모델의 기초 (Margarethe McDonald)

Linear mixed effects model basics 

This session will cover preparing a dataset, building and running maximal models, dealing with model convergence issues, and checking model assumptions using the lme4 package in R. 

Afternoon session

• 13:00 - 14:00 | Tutorial Session 2: 선형 혼합 효과 회귀 모델의 구축및 결과 해석 (Margarethe McDonald)

Interpreting linear mixed effects model results 

This session will cover extracting model results, doing follow-up analyses using the emmeans package, and visualizing using ggplot2.

• 14:00 - 14:30 | Research talk 2: 질적/범주적인 변수를 종속변수로 하는 연구 소개 (고언숙)

How do mothers and children initiate conversational exchanges?: The dynamics of multimodal cue usage in beginning vocal exchanges across child development

• 14:30 - 15:30 | Tutorial Session 3: 데이터 변환과 선형 혼합 효과 회귀 모델 (Jun Ho Chai)

Data transformation and linear mixed models 

This session introduces a multimodal dataset of mother-child interactions with annotations of non-verbal behavioral cues, temporally segmented relative to speech events. We will cover basic data transformation to convert qualitative & categorical data to frequency and proportion to address different research questions. We will then fit LMM using lme4 package and inspect the model.

• 15:30 - 16:30 | Tutorial Session 4: 상호작용의 해석: 주변평균, 회귀, 그리고 시각화 (Jun Ho Chai)

Interpreting interactions: Marginal means, Regressions, and Visualization 

We will explore interactions between categorical variables and compare their estimated marginal means using the emmeans package. We will learn how to extract marginal regressions and trends using the modelbased package. We will go through a straightforward process to visualise the effects using the ggeffect package.